Semi-supervised feature selection
(2013)
Files
20130310mainTese2-final.pdf
Open Access - Adobe PDF
- 7.74 MB
Details
- Authors
- Supervisors
- Abstract
- (en) As data acquisition has become relatively easy and inexpensive, data sets are becoming extremely large, both in the number of variables and in the number of instances. However, the same is not true for "labeled" instances. Usually unlabeled data represent the majority of instances. Using such data requires special care, since several problems arise with the dimensionality increase and the lack of labels. Reducing the size of the data is thus a primordial need. In this context, the application of a semi-supervised approach is very suitable, where one can try to take advantage of the best benefits that each type of data has to offer. The problem can be addressed in the context of feature clustering, grouping similar variables, or through a multi-objective approach, since we have arguments that clearly establish its multi-objective nature. In the first approach, a similarity measure based on mutual information, capable to take into account both the labeled and unlabeled data, is developed. The principle of homogeneity between labels and data clusters is also exploited and two semi-supervised feature selection methods are developed. Finally a mutual information estimator for a mixed set of discrete and continuous variables is developed as a secondary contribution. In the multi-objective approach, the proposal is try to solve both the problem of feature selection and function approximation, at the same time. The proposed method includes considering different weight vector norms for each layer of a MLP neural network, the independent training of each layer and the definition of objective functions that are able to eliminate irrelevant features.
- Affiliations
Citations
Gualberto Ferreira Coelho, F. (2013). Semi-supervised feature selection. https://hdl.handle.net/2078.5/205457